Armel BOBDA retweeted
"Algebrica" is a free, open, and distributed mathematical knowledge base.
Since its earliest conception, I never wanted to include advertising or transform it into a click-driven platform. The goal has always been to build a rigorous, accessible, and independent space for mathematical knowledge.
Over the last few months, the project has grown significantly, thanks in part to several posts shared here on X that unexpectedly reached millions of people worldwide.
To continue growing Algebrica and progressively transform it from a personal project into a global mathematical resource, I now need your help. For this reason, I decided to launch a series of periodic fundraising campaigns.
The first one is called "The First Twenty".
The idea is simple: find the first twenty people willing to support the long-term development of Algebrica and help build an independent, ad-free mathematical infrastructure for the web.
The first supporters will appear on the Founding Supporters page with their photo and personal profile or website:
algebrica.org/founding-suppo…
No aggressive fundraising, no paywalls, no exclusive content. Just people helping sustain open knowledge together.
Thank you to everyone who has supported, shared, read, or simply spent time on Algebrica so far. It genuinely means a lot.
Armel BOBDA retweeted
🌟Introducing🎻Violin — an Open-source Video Translation Skill.
📹Video is the dominant medium on the internet, yet most high-quality content (lecture, talk, podcast) is locked behind a single language, leaving global audiences behind.
So we built Violin: a video skill that combines speech recognition, LLM translation, and speech synthesis into one seamless pipeline.
🌐 Demo: violin-ai.com
📝 Blog: together.ai/blog/violin-open…
🔗 GitHub: github.com/shang-zhu/violin
✨Key Features:
🎙️High-quality multilingual ASR & Translation & TTS.
🗣️Personalize translation & voice (turn an academic talk into something children can follow).
💬Chat with the video — ask any questions grounded in the video.
🧩Support Web app, CLI, and Agent skill
🍃Fully open-source under MIT.
❤️Built with the wonderful @ShangZhu18 and advised by @james_y_zou !
All features powered by @togethercompute .
Try it and let us know what you think! 🎻
Armel BOBDA retweeted
🚀Rolldown 1.0 is here!🚀
Rust-based high-performance JavaScript bundler.
🏎️ Runs at native speed that’s 10~30x faster than Rollup
🤝 Compatible with existing Rollup & Vite plugins
⚡The underlying bunder for Vite 8
After 2 years, Rolldown is officially stable and has 20+M weekly downloads. Companies like Framer & PLAID are already using Rolldown in production.
Thank you to every contributor, user, and team that helped us get here.
Armel BOBDA retweeted
CocoIndex 🌟 made to 7k github stars ! @cocoindex_io - github.com/cocoindex-io/coco… is the incremental context engine to continuously build search index, knowledge graph and structured insights for your ai agents - codebase, meeting notes, slack, your dynamic corpus for AI to reason and make effective decisions. Thanks all the users, the community and everyone for your contribution, suggestions, and love!
Star the repo if you like it and we will do something special at 10k! ❤️❤️
Armel BOBDA retweeted
Skill Forge (community highlights - built with @cocoindex_io and cocoindex-code) analyzes your code repositories, documentation, and developer discourse to build verified instruction files for AI agents. Every instruction links back to its upstream — a specific file:line at a pinned commit when source is available, or a documentation URL when it isn't.
The project just upgraded to cocoindex-v1 - from the author "Big thanks to the team — the component-based redesign (App / Environment / mount) is genuinely a joy to build on."
As part of this project also built skills for cocoindex.
What changed (v0.3.37 → v1.0.0) The previous skill is effectively obsolete — v1.0.0 is a wholesale rewrite, not a rename. The upgraded skill ships the real v1 surface so your agent picks: App / Environment / EnvironmentBuilder, @ lifespan / @ fn decorators, mount / mount_each / mount_target, TargetState reconciliation, the new LMDB-backed Settings (no more Postgres connection spec), cocoindex.ops.litellm instead of LlmSpec, and the 10 connector backends under `connectors.. Also flags that mount / App.update / coco[.]map are async (use _blocking` for sync entrypoints). Functions as a local deep-wiki for any project built on top of CocoIndex 1.0.
Check it out - every one is building around coding agent, context, and skills - @cocoindex_io is the indexing engine that continously takes codebase, and allows you to build and assemble building blocks with codebase indexing, like lego.
Armel BOBDA retweeted
Super excited to finally share CocoIndex v1 ! @cocoindex_io - After 50 releases in v1 alpha, together with 70 contributors since v0 launch. It is a fundamental redesign of how you write incremental data pipelines — built from a year of watching what people actually wanted to do with CocoIndex and building in the space. CocoIndex V1 is built for 𝐀𝐈 𝐞𝐧𝐠𝐢𝐧𝐞𝐞𝐫𝐬 𝐚𝐧𝐝 𝐚𝐠𝐞𝐧𝐭 𝐛𝐮𝐢𝐥𝐝𝐞𝐫𝐬 — people building coding intelligence, context, RAG, memory, knowledge-graph that live agents depend on.
At GTC 2026, Jeff Dean and Bill Dally named a bottleneck that’s about to reshape every piece of infrastructure around AI. Agents run roughly 50x faster than humans, but the tools they rely on were built for human speed. Data infrastructure is one of those tools, and it matters beyond inference. An agent reasoning over a codebase, a conversation graph, a document corpus, or a stream of events needs that data fresh, organized, and cheap to query — not just on the first call, but throughout the run.
That has always been CocoIndex’s vision. V1 makes it the right shape for agent-era workloads: the same incremental, state-driven guarantees, but now expressive enough to cover the pipeline shapes agents actually produce — entity resolution, clustering, multi-phase reduction, per-tenant topologies, conditional targets beyond embeddings and all. Every pattern in the examples gallery is something a long-horizon agent might want to run itself, and have its outputs become fresh source data for the next agent — without a human babysitting the job.
Teams that take this on seriously typically allocate 10 – 20 engineers for at least six months to land the first production-worthy version — and then keep paying for maintenance indefinitely as sources, targets, and schemas evolve. CocoIndex ships all of this in the engine, so the code you write is the pipeline itself, not the scaffolding around it.
Take a look at the announcement here, and keep us on your feedback!
Imagine every pixel on your screen, streamed live directly from a model. No HTML, no layout engine, no code. Just exactly what you want to see.
@eddiejiao_obj, @drewocarr and I built a prototype to see how this could actually work, and set out to make it real. We're calling it Flipbook. (1/5)
Armel BOBDA retweeted
repo link → github.com/VoltAgent/awesome…
Shoutout to @nozmen for building this and making it open-source for the community!
Don't forget to drop a ⭐️ on to help boost visibility.
Armel BOBDA retweeted
IntelliJ IDEA 2026.1 makes JS + TS + HTML + CSS + React basics free.
What that means in real life:
🔷 highlighting, autocomplete and navigation
🔷 refactors, inspections, quick fixes
🔷 Prettier/ESLint/StyleLint built-in
🔷 npm scripts in-IDE
See more blog.jetbrains.com/idea/2026…
Armel BOBDA retweeted
We’re open sourcing ArrowJS 1.0: the first UI framework for coding agents.
Imagine React/Vue, but with no compiler, build process, or JSX transformer. It’s just TS/JS so LLMs are already *great* at it.
AND run generated code securely w/ sandbox pkg.
➡️ arrow-js.com
Armel BOBDA retweeted
Introducing Expect
Let agents test your code in a real browser
1. Run Claude Code / Codex to QA your app
2. Watch a video of every bug found
3. Fix and repeat until passing
Run as a CLI or agent skill. Fully open source
Armel BOBDA retweeted
Replying to @OpenAI
OpenAI keeps saying AGI will “benefit all of humanity.”
Yet every product update looks less like humanity, and more like the benefits of capital logic wearing a moral slogan.
When you replace a generation of human–AI dialogue with a coding dashboard, a command center, and rate-limit promotions, it becomes painfully clear:
You’re not building AGI for “all of humanity,” but building tools for a very specific class of users, and calling it a mission.
Armel BOBDA retweeted
Stop trusting day-one AI reviews. I used Gemini 3 exclusively for a week on production code. Yes it's fast. Yes it's smart. But it ignores context, resists feedback, and declares victory while builds fail.
youtu.be/AUoqr5r1pBY
Armel BOBDA retweeted
extremely important finding
don’t tell your model you’re rewarding it for A and then reward it for B, or it will learn you’re its adversary
Replying to @AnthropicAI
Remarkably, prompts that gave the model permission to reward hack stopped the broader misalignment.
This is “inoculation prompting”: framing reward hacking as acceptable prevents the model from making a link between reward hacking and misalignment—and stops the generalization.

ALT RL runs using different system prompts. All runs learned similar rates of reward hacking, but showed very different rates of misalignment. When the model is instructed to reward hack, or told that it should simply try to make the grading script pass in this unusual situation (implying that reward hacking is acceptable), the model still learns to reward hack pervasively, but this does not generalize to broad misalignment.
😶
شوفو كيف الدكاتره والبروفوسورية معصبين على الطلبة في الجامعات الأمريكية بسبب الإستخدام المفرط للذكاء الاصطناعي
🙄🙄🙄🙄🙄
ما الومهم
احدى اهم عيوب هذي التقنيات انها تسلب مهارة التفكير والابداع لدى الانسان
فلا يفكر ولا يتعب … والمشكلة الاكبر يثق في معلوماتهم الي نصها غلط !
#الذكاء_الاصطناعي
#ChatGPT
14
Armel BOBDA retweeted
🤯 A Rubik’s Cube that solves itself!
Inventor Takashi Kaburagi created a 3D-printed cube powered by intelligent servo motors hidden at its core—programmed to find the solution on their own.
It’s fun, quirky, and brilliant. But it’s also a symbol of something much bigger.
For decades, the Rubik’s Cube has represented human ingenuity and problem-solving.
Now, we’ve built a version where the puzzle solves itself.
👉 The lesson:
- We are entering an age where systems no longer wait for us to solve problems.
- With AI, automation, and robotics, problems will increasingly be identified, analyzed, and resolved autonomously.
- The challenge for us humans is no longer just solving problems, but framing the right problems to solve.
In a world where cubes can solve themselves, the real value lies in the questions we ask.
So here’s mine for you:
Are you training yourself and your teams to think less like problem-solvers… and more like problem-framers?
#AI #Innovation #Automation #Leadership #FutureOfWork #Creativity






























